Auto-updating model-based control for thrust variation mitigation and acceleration performance enhancement of gas turbine aero-engines

Abstract

Model-based control shows promising potential for engine performance improvement and future aero-propulsion requirements. In this paper, an auto-updating thrust variation mitigation (AuTVM) control approach using on-board model strategies is proposed for gas turbine aero-engines under in-service degradation effects, which aims at active thrust regulation and acceleration protection in a simultaneous way. The AuTVM control is integrated with an on-line block, based on a reliable on-board engine model, and an off-line part for the periodical update of control parameters via post-flight engine monitoring data. The core feature of the AuTVM control is a set of auto-updating loops within the on-line part, including thrust regulation loop, surge margin loop, turbine entry temperature loop, and the steady loop, whose control parameters are periodically adjusted with increasing flight cycles. Meanwhile, an industrial sensor-based baseline controller and two tailored model-based controllers, i.e., a thrust variation mitigation (TVM) controller with fixed gains and a self-enhancing active transient protection (SeATP) controller with pro-active transient protection and passive thrust control, are also developed as comparison bases. Numerical simulations for idle to full-power acceleration tests are carried on a validated aero-thermal turbofan engine model using publicly available degradation data. Simulation results demonstrate that both new engines and severely degraded engines regulated by the AuTVM controller show significant thrust response enhancement, compared to the baseline controller. Moreover, thrust variation at the maximum steady state of degraded engines, which exists within the SeATP controller and the baseline controller, is suppressed by the proposed AuTVM controller. Robustness analysis against degradation uncertainties and sensor accuracy confirms that the AuTVM controller owns a closer maximum steady-state thrust distribution to the desired value than those of the SeATP and the baseline controller while utilizing transient margins of controlled engines more effectively. Hence, the control performance of the AuTVM controller for in-service engines is guaranteed.